Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 16222
    Prev
    Next
    INDEX
    Explanations

    political party names, especially the Republican Party

    oai_token-act-pair · gpt-3.5-turbo

    mentions of political parties

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     composite
    -0.71
    aer
    -0.68
    eval
    -0.68
    ebook
    -0.65
     Archangel
    -0.64
    eus
    -0.62
     McMaster
    -0.60
    zsche
    -0.60
    gomery
    -0.60
    orest
    -0.60
    POSITIVE LOGITS
     Party
    3.88
    Party
    2.74
     Parties
    2.65
     party
    2.63
    party
    2.42
     parties
    1.79
     Caucus
    1.25
     caucus
    1.12
     PART
    1.12
     Polit
    1.05
    Activations Density 0.014%

    No Known Activations